Rafn
UW-L Journal of Undergraduate Research XII (2009)
How to Stay Healthy? An Exploration in Local Attitudes Towards U.S.
Health Care Policy
Gregory Rafn
Faculty Sponsor: Carol Miller, Department of Sociology and Archaeology
ABSTRACT
Currently in the United States there are approximately 47.5 million Americans without health
insurance (Urbina 2008). Previous research has attempted to tackle this growing social problem
by investigating alternative health care policies and the effects they might have on various
populations within the United States. This study seeks to investigate local residents’ attitudes
pertaining to health care policy in the United States and to determine the extent to which these
attitudes of a Midwestern city of approximately 50,000 people vary by socioeconomic status.
Utilizing survey methodology I investigate potential relationships between variables such as
income, age, gender, and employment status and sentiments on universal health care, privatized
health care, and inequality in the health care system in a geographical region where health care
costs have risen faster than the national average. Furthermore, I test whether a significant age,
gender and socioeconomic difference on attitudes about health care exist.
INTRODUCTION
There are 47.5 million Americans without health insurance (Urbina 2008). The high number of uninsured
citizens, skyrocketing health care costs while wages decline, and the increasing transfer of such costs from
employers to employees are just a few of the many problems with health care in the United States (Singer 2007). To
date, the United States spends twice as much per person as Canada on health care, yet Canadians have a longer life
expectancy (Clarke 2008). Individuals with health insurance must often pay high deductibles before insurance takes
effect (Hoffman 2006). This type of insurance deliberately discourages individuals from overusing health care
services (Hoffman 2006). Furthermore, inequalities in health care services further intensify socioeconomic
inequalities in the United States (Singer 2007).
Many plans for overhauling the American health care system exist. Laurence Kotlikoff (2007) proposes a 10point Medical Security System that would require mandatory participation and award vouchers to all who purchase
health insurance. This system would base vouchers off of individual experience and would provide more resources
to high health risk individuals (Kotlikoff 2007). David Singer (2007) proposes a single-payer health coverage plan
in which all private health insurance plans are eliminated and replaced with a Canadian-style system paid through
payroll taxes. Each unique plan involves a major overhaul of our current health care system, affecting numerous
individuals, rich and poor, but do we really know which system is best for everyone?
Attitudes on how the government should address health care have fluctuated over time. As of November of
2007, 64% of Americans felt that it was the government’s responsibility to provide health care access to all citizens
(Gallup 2007). In a January report, Americans were asked whether they would be in favor or oppose a universal
health care plan. Of those polled, 62.8% favored a universal health care policy (Blendon, Altman, Deane, Benson,
Brodie, and Buhr 2008).
This issue is experienced locally, as well. La Crosse County’s health care costs are increasing and have more
than doubled since 1999 (Magney 2007). As a result, local officials are looking towards an alternative method for
funding health care. These methods include experimenting with differing levels of deductibles and implementing
prescription cards (Magney 2007).
In order to better understand what changes to our current health care system would be beneficial I investigated
whether certain demographic variables such as age, gender, income, education, and household size influence
individuals’ attitudes towards their confidence in their ability to afford health care and their openness to government
intervention within health care. I hypothesized that income would be the primary variable influencing an
individuals’ confidence in his or her ability to afford health care. I further hypothesized that individuals with higher
educational attainment would be more in favor of government intervention with in the health care system.
1
Rafn
UW-L Journal of Undergraduate Research XII (2009)
METHODS
In order to effectively and efficiently survey La Crosse resident’s I created an approximately 6-9 minute phone
survey to obtain residents’ attitudes and opinions regarding various health care policy. I used the Survey Research
Center (CWH 323) available to sociology students at UW-L in order to conduct each survey. The sample size of
100 was compiled utilizing a systematic random sampling method to ensure that a probability sample was acquired.
Using this method I randomly selected phone numbers from a current La Crosse area phone book simply by opening
the page and selecting a starting point for calling. From this starting point every 15th number was subsequently
chosen to participate in the survey. Businesses and residents’ with an address listed outside of La Crosse were
omitted. Upon calling each number a brief introduction was given to explain the purpose and importance of this
study. Individuals were made sure to be over the age of 18 and were notified that their responses would remain
confidential and that participation may be terminated at any point. After I presented all of this information,
individuals gave their verbal consent to the survey terms and we began the survey.
The survey consisted of several different types of questions. I used questions derived from two scholarly
sources. The Gallup Poll, a behavioral and opinions research organization, supplied my survey with essential
questions revolving around opinions pertaining to universal versus privatized heath care policies. For example, one
question used from this organization was “Do you think it is the government’s responsibility to make sure all
Americans have health care coverage, or is that not the responsibility of the federal government?” Another question
utilized was, “Which of the following approaches for providing health care in the United States would you prefer:
replacing the current health care system with a new government run health care system, or maintaining the current
system based mostly on private health insurance?” Both of these questions provided a foundation to begin
investigating attitudes towards current health care policy.
The second source from which survey questions were adapted was a 2008 Health Confidence Survey created by
the Employee Benefit Research Institute. I used questions that targeted specific facets of health care and strategies
to either reduce costs or increase the amount of individuals eligible or capable of obtaining some form of health
insurance. Typically these questions used a Likert scale to assess specific attitudes on each ranging from Strongly
Agree/Support to Strongly Disagree/Oppose. One specific question focused on whether individuals felt it was their
employers’ responsibility to provide them with health insurance. Questions also revolved around the individuals’
confidence in the current health care system as well as their ability to afford necessary health services. This specific
question asked respondents to rate their confidence in their ability to afford health care services. Responses ranged
from Extremely Confident to Not At All Confident.
The last seven questions in the survey were basic demographic questions that pertained to the respondents’ age,
gender, race, income, employment, household size, and highest year of education received. These questions were
placed at the end of the survey to avoid the situation of individuals opting out of participating in the survey because
they were reluctant to divulge sensitive information.
RESULTS
Initial descriptive statistics were run to establish the demographics of the sample population. Frequencies were
run to give an indication of the number of individuals in each education level, their annual income, and their gender.
Descriptive statistics to establish the average age of respondents was also run. The variables: education level, age,
and income were all recoded due to the wide disparity between the responses. All variables, excluding gender, were
recoded into dichotomies, so they could be included as dummy variables in various analyses. Age was recoded into
Younger than 65 and 65 and Older. Education level was recoded into High School or Less and Some College or
More. Annual income was recoded into Less than $25,000 and Greater than $25,000. Tables 1 through 4 indicate
the frequency of responses pertaining to each variable.
Tables 1-6. Descriptive Statistics for La Crosse Residents
Table 1. Age
Variable
Age
N = 89
Mean
63.82
2
Standard Deviation
15.57
Minimum
19
Maximum
92
Rafn
UW-L Journal of Undergraduate Research XII (2009)
Table 2. Gender
Male
Female
Total
Frequency
42
58
100
Percent
42.0%
58.0%
100%
Valid Percent
42.0%
58.0%
100%
Frequency
38
62
100
Percent
38.0%
62.0%
100%
Valid Percent
38.0%
62.0%
100%
Frequency
28
61
89
Percent
31.5%
61.0%
89.0%
Valid Percent
31.5%
68.5%
100%
Table 3. Education Level
High School or Less
Some College or More
Total
Table 4. Annual Income
Less than $25,000
Greater than $25,000
Total
Descriptive statistics pertaining to the number or percent of La Crosse residents that are satisfied or dissatisfied
with current health care costs as well as their perception on the quality of health care coverage were run to establish
a basis of residents’ opinions on the current health care policy. The variable pertaining to health care coverage in
the United States was recoded into two categories: Good and Poor respectively. Results indicated that 70.0% of
respondents are currently dissatisfied with overall health care costs in the United States. Furthermore, 48.0% of La
Crosse residents surveyed specified that health care coverage in the United States is poor. The frequency of the
responses to each of these variables is specified in tables 5 and 6.
Table 5. Descriptive Statistics for Satisfaction of Health Care Costs
Frequency
Satisfied
20
Dissatisfied
70
Total
90
Percent
20.0%
70.0%
90.0%
Valid Percent
22.2%
77.8%
100%
Table 6. Descriptive Statistics for Perception of Health Care Coverage
Frequency
Poor
48
Good
51
Total
99
Percent
48.0%
51.0%
99.0
Valid Percent
48.5%
51.5%
100%
Two individual Chi-square tests were run to investigate any significance between a respondent’s confidence in
his or her ability to afford health care and his or her annual income. In the first test age was used as a control since
age was found to be an influence on the effects of income on the respondent’s confidence. The results show a
significant relationship for individuals younger than 65 with a X2 = 10.040 and a p-value less than .05; (p =. 007.)
The results for individuals older than 65 are nearly significant, however, do not fully establish a relationship with a
Χ2 = 5.880 and a p-value slightly greater than .05; (p =. 053.) The results are shown in table 1 below.
Table 1. Perception of Respondent’s Ability to Afford Health Care by Income
Less than $25,000
Greater than $25,000
Younger than 65 Years
3
Total
Rafn
UW-L Journal of Undergraduate Research XII (2009)
Very Confident
10.0%
Somewhat Confident
20.0%
Not at all Confident
70.0%
Total N
10
%
100%
Χ2 = 10.040, df = 2, p = .007, C = .435, Φ = .483, λ= .100
Less than $25,000
65 Years or Older
Very Confident
27.8%
Somewhat Confident
66.7%
Not at all Confident
5.6%
Total N
18
%
100%
45.5%
36.4%
18.2%
33
100%
37.2%
32.6%
30.2%
43
100%
Greater than $25,000
64.3%
32.1%
3.6%
28
100%
Total
50.0%
45.7%
4.3%
46
100%
Χ2 = 5.880, df = 2, p = .053, C = .337, V = .358, λ= .167
The second Chi-square test examined the same relationship, however, education level was used as a control
since educational level also influences the effects of income on the respondent’s confidence. The results show a
significant relationship with a Χ2 = 7.778 and a p-value less than .05; (p =. 020) for individuals with an education
level of high school or less. For individuals with some college or more the results were also significant with a Χ2 =
9.956 and a p-value less than .05; (p =.007.) The results are shown in table 2 below.
Table 2. Perception of Respondent’s Ability to Afford Health Care by Income
Less than $25,000
Greater than $25,000
High School or Less
Very Confident
16.7%
56.5%
Somewhat Confident
75.0%
26.1%
Not at all Confident
8.3%
17.1%
Total N
12
23
%
100%
100%
Χ2 = 7.778, df = 2, p = .020, C = .426, V = .471, λ= .250
Less than $25,000
Greater than $25,000
Some College or More
Very Confident
25.0%
52.6%
Somewhat Confident
31.2%
39.5%
Not at all Confident
43.8%
7.9%
Total N
16
38
%
100%
100%
Χ2 = 9.956, df = 2, p = .007, C = .395, V = .429, λ= .250
Total
42.9%
42.9%
14.3%
35
100%
Total
44.4%
37.0%
18.5%
54
100%
A regression model was run to test which variables influenced individuals’ attitudes about government
intervention within health care. An index was created using survey questions that focused on whether it was more
of a government responsibility to make sure citizens have adequate health care or whether it was up to the individual
to seek resources to ensure adequate health care. The index was tested against each respondent’s education level,
annual income, their satisfaction of health care costs, age, the number of adults in their household, and gender. The
results concluded three significant variables, education level, respondent’s satisfaction of health care cost, and the
number of adults living in their household with p-values less than .05, attributed to influencing whether an
individual is more likely to support government intervention in health care. The results are shown in Table 1 below.
Table 1. Regression of La Crosse Resident’s Attitudes of Government
Intervention in Health Care.
Variables
Unstandardized
Coefficient
4
Rafn
UW-L Journal of Undergraduate Research XII (2009)
Constant
Education
Level
Annual
Income
Satisfaction
of Health
Care Costs
Age
# of Adults
Living in
Household
Gender
Adjusted R2
N
(Standard Error)
13.499
(2.913)
1.853*
(.830)
-.672
(.887)
-2.476*
(1.002)
.808
(.855)
1.180*
(.576)
-1.366
(.786)
.125
69
A Chi-square test was run to investigate a relationship between a respondent’s attitude whether it was the
government’s responsibility to ensure health care coverage and their annual income. The results concluded a
significant relationship with a Χ2 = 4.415 and a p-value less than .05; p =.036. The results are shown in the table
below.
Table 1. Perception of Government’s Responsibility in Regards to Providing Health Care by Income.
Less Than $25,000
Greater Than $25,000
Yes, Government
74.1%
50.0%
Responsibility
No, not Government
25.9%
50.0%
Responsibility
Total N
27
60
%
100%
100%
Χ2 = 4.415, df = 1, p = .036, C = .220, V = .225, λ= .000
Total
57.5%
42.5%
87
100%
A Chi-square test was run to investigate a relationship between a respondent’s opinion on government
intervention in health care based off of whether he or she supported the government requiring all employers to pay
towards health care insurance and their education level. The results concluded a significant relationship with a Χ2 =
10.249 and a p-value less than .05; p= .017. The results are shown in the table below.
Table 1. Perception of Employer’s Responsibility to Pay Towards Health Insurance by Education Level.
High School or Less
Some College or More
Strongly Support
62.2%
32.3%
Somewhat Support
21.6%
24.2%
Somewhat Oppose
8.1%
27.4%
Strongly Oppose
8.1%
16.1%
Total N
37
62
%
100%
100%
Χ2 = 10.249, df = 3, p = .017, C = .306, Φ = .322, λ= .081
5
Total
43.4%
23.2%
20.2%
13.1%
99
100%
Rafn
UW-L Journal of Undergraduate Research XII (2009)
DISCUSSION
The initial descriptive statistics provide a strong foundation to continue research around this topic. La Crosse in
the past 2 years has seen an exponential growth in health care costs and insurance premiums. (Magney 2007). My
research findings reveal that local residents, especially elderly individuals, are beginning to experience the burden
these costs have on their daily life and influence everyday decisions. With 70% of individuals dissatisfied with
health care costs and almost 50% of individuals identifying that there are serious gaps in health care coverage
nationally, it is apparent that negotiations over new health care legislation and policies is greatly needed not only
locally but also nationally.
My initial hypothesis of income being the most influential variable on an individual’s confidence in the ability
to afford health care coverage is supported by my significant findings. Results indicated that when controlling age,
individuals with an estimated annual income of less than $25,000 held very little confidence in their ability to afford
today’s health care costs. Individuals with an estimated annual income greater than $25,000 were more confident in
their ability to afford health care costs. By comparing the two age groups, individuals younger than 65 are
significantly less confident in their ability to afford health care compared to individuals older than 65. This can be
attributed to the unpredictability of health care coverage and costs in the future that younger generations are more
likely to experience. Furthermore, individuals 65 years and older are eligible for Medicare and receive some form
of continuous assistance towards health care costs. This in turn influences older individuals opinions on the costs of
health care since there is some continuity in the benefits received.
Results also indicated that when controlling education level, a respondents’ annual income continued to
significantly influence their confidence in their ability to afford adequate health care services. In this instance,
however, individuals with some college or more earning less than $25,000 significantly reported less confidence in
their ability to afford health care when compared to individuals who had completed high school or less earning less
than $25,000. Due to the demographics of the data collected, I suspect that the majority of individuals who
responded as having completed high school or less are older in age. Therefore, during their lifetime a high school
education provided them with the necessary skills and abilities to obtain a relatively well paying job with adequate
benefits. Now older, these individuals either maintain some benefits from the time period in which they worked in
or are eligible for Medicare benefits.
Individuals with some college or more are most likely younger respondents below 65 years of age. In today’s
society and faltering economy a college education often with graduate work is now necessary in order to obtain a
well paying career with sufficient benefits. These individuals understand the difficulty to obtain such a career and
the costs necessary to receive sufficient education. This in turn hinders their confidence in the ability to afford
health care coverage, especially with dramatically rising costs.
My second hypothesis was supported by the results. Education level, the respondent’s satisfaction of health
care costs, and the number of adults living in their household significantly influence a respondent’s willingness to
allow government intervention in the health care system. The results indicate as education level increases the
individual is more likely to allow government intervention. This can be attributed to an increase in an individuals’
understanding of both government policies/legislation as well as the overall mechanics of the health care system.
Education level also influences an individual’s opinion on whether employers should be required by the government
to pay towards health insurance for employees. Individuals with a lower income are more likely to be employed in
low level service or technical trades that do not offer health insurance benefits therefore, making it extremely
difficult to maintain an adequate standard of living and the ability to afford health care services.
As a respondent’s satisfaction in the overall cost of health care increases they are more likely to oppose
government intervention. Individuals who are satisfied with health care costs today either have adequate health
insurance coverage and/or finances to ensure continuous coverage in future years.
As the number of adults living in the same household increases respondents are more willing to allow
government intervention in health care. Clearly due to continually rising health care costs it is harder and often
more expensive to insure multiple individuals within the same household. Allowing government intervention or
changes to our current health care system brings hope the lower costs and expansions in coverage will occur.
Income also significantly influences whether an individual feels it is the government’s responsibility to ensure
health care coverage for all citizens. Individuals reporting an annual income of less than $25,000 were more likely
to indicate that the government had a responsibility in ensuring health care coverage. This I feel is related back to
their confidence in their ability to afford health care. Individuals earning less than $25,000 are at a disadvantage
with less disposable income available for medical services. I suspect that lower income individuals hope a change in
health care will bring more financial assistance towards health care costs.
6
Rafn
UW-L Journal of Undergraduate Research XII (2009)
LIMITATIONS
Certain limitations present themselves throughout this study and need to be addressed in future research. The
use of phone survey methodology is an inexpensive and convenient way to administer surveys to a broad population
in a relatively short time period. In today’s era, however, not all individuals utilize landline phones as their primary
phone service. The result is all individuals using cellular phones as their primary phone service are completely
missed and not given the opportunity to express their opinions in this survey. I feel that the increase in use of
cellular phones, especially among younger generations resulted in an important group of individuals throughout the
La Crosse community were neglected from participation in the survey.
The availability of the UW-La Crosse Sociology Call Center is one other limitation in this study. Times that
were open for calling were predominately during the mid-day hours. This resulted in obtaining a skewed sample
population due to the majority of participants were either retired or stay at home parents. It is important to note that
this population is important to consider in research, however, the over representation of this population in the sample
resulted in a limited variance among opinions and attitudes toward health care policy.
ACKNOWLEDGEMENTS
I would like to thank Carol Miller for all her assistance throughout this project. I would further like to thank
Kayla Breckheimer, Angela Rauchwarter, and Stephanie Weist for their assistance in calling and administering
surveys. I would also like to thank the UW-L Undergraduate Research Grants Program for providing me with the
opportunity to conduct this project.
REFERENCES
Blendon, Robert J., Drew E. Altman, Claudia Deane, John M. Benson, Mollyann Brodie, and Tami Buhr. 2008.
“Health Care in the 2008 Presidential Primaries.” New England Journal of Medicine 358: 414-422.
Clarke, Rory. 2008. “How Healthy is our Health Care?” OECD Observer, 2008.
http://www.oecdobserver.org/news/fullstory.php/aid/629/How_healthy_is_ou_healthcare_.html. Retrieved
September 29, 2008.
Gallup Poll. 2007. “Health Care System in the United States.” Gallup Poll.
http://www.gallup.com/poll/4708/Healthcare-System.aspx. Retrieved September 29, 2008.
Hoffman, Beatrix. 2006. “Restraining The Health Care Consumer.” Social Science History 30, 4:501-528.
Kotlikoff, Laurence J. 2007. “The Healthcare Fix: Universal Insurance for All Americans.” Cato Journal 28,1: 155159.
Magney, Reid. 2007. “County Struggles with Health Insurance Costs.” La Crosse Tribune Online, October 16.
http://www.lacrossetribune.com/articles/2007/10/16/news/z01countybudget.txt. Retrieved September 29, 2008.
Singer, David. 2007. “The Health Care Crisis in the United States.” Monthly Review 59,9: 18-29.
Urbina, Ian. 2008. “A Decline in Uninsured Is Reported for 2007.” New York Times, August 26.
http://www.nytimes.com/2008/08/27/washington/27census.html?scp=1&sq=uninsured20americans%202008&s
t=cse. Retrieved September 29, 2008.
7